The Problem with the Marshmallow Test

Resilience (Part 2) – Types of Resilience
May 1, 2020
Resilience (Part 3) – False Sense of Resilience
May 25, 2020

Whether you are a psychology student or not, you probably have heard about the Marshmallow test! The benefits of this interesting test are said to be far-reaching, so much so that parents may even try it with their young – hoping they would not be disappointed. So what is the fuss all about?


The Marshmallow Test

The Marshmallow test dates back to the 1960s and 1970s in the original research conducted by Stanford University psychologist Walter Mischel and his colleagues. Children between three and five years old were given a marshmallow that they could eat immediately or resisted eating for 10 minutes. If they resisted, they would be rewarded with two marshmallows. 

The same researcher and his colleagues conducted a follow-up study with the same subjects in 1990, some 20 years later. And the results were astounding! Compared to those who gave in to temptation, the children who could delay gratification exhibited advanced traits of intelligence and behaviour in their late teenage years. Specifically, the researchers found that they achieved higher SAT scores, better emotional coping skills and healthier body weights.

The profound, tangible benefits of delayed gratification on outcomes later in life was a huge hit with parents. Anxious parents began exploring ways and means to improve their child’s ability to delay gratification. This happened in the decades since Mischel’s work on the Marshmallow test sent waves in the field of educational psychology and permeated parenting advice among the middle-class.

I am sure you will not be surprised if parents have a checklist of the following stashed away somewhere within easy reach:

After putting my child through a certain program, does my child:

  • Yield less to temptation?
  • Get less distracted when trying to concentrate?
  • Become more intelligent?
  • Become more self-reliant?
  • Become more confident?
  • Become more able to trust their own judgment?

The list goes on…


Issues with the Original Research

Clearly, there are many in society who have taken action based on the staggering findings from the original research. However, others have raised concerns mainly in two key areas. Some questioned how the researchers made causal connections between the child’s ability to delay gratification /impulse control for eating the marshmallow and their choices later in their teenage lives, without controlling for other variables. For instance, the social and economic background of the participants.

Another area of concern had to do with the research conclusions that were based on a small sample size, including only children of the Stanford University staff. These children went to preschool on the university campus, which limited the pool of participants to the offspring of professors and graduate students.

Even for the follow-up study that was conducted a decade later, it lacked consistency as it included fewer than the 50 children of the sample Mischel and colleagues formed their conclusions.



A New Research

As with other reputable studies, this gave rise to a replica of the study in a new research by Tyler Watts, Greg Duncan and Hoanan Quen. However, this time with marked adjustments. 

The new research had a sizable and more representative sample size of 900 children. It included data on preschoolers whose parents did not have 

college degrees, along with those whose parents had more higher education. It also included a more diverse array of individuals with a range of ethnicities and income levels.

Additionally, the researchers made side by side comparisons between children who were similar in all of the following ways: ethnicity, gender, type of home environment, type of parents, general cognitive ability, measured very early on. The only difference that set them apart was their ability to delay gratification.

It is 2018, and published in the Psychological Science journal was a surprising discovery. While children still enjoyed benefits of delayed gratification, the effects in the revisited test were far less significant than those found by Mischel and his team in the original research. In fact, the effects even disappeared to a large extent for subjects at 15 years old, after they took into account family and parental education.

Given these results, it was a decisive message to parents, schools and nurseries. Coaching children to delay gratification will not lead to all the expected magic in their adolescence. It also means that we need to take a closer look at those who succeeded and waited longer for a greater reward. Was it because of an innate quality they had? Or was it something more external?


The Plot Thickens

Further findings revealed that those who could wait longer also had another common characteristic – they mostly came from affluent households. As such, their future success appears to be based more on this socio-economic advantage rather than sheer willpower. They typically do not need to worry about material provision. Therefore, it is easier to delay gratification because their experience assures them that their family has the means to provide later on, if not now.

With that, it is interesting to consider that the original marshmallow test may have revealed other limitations of the findings. The extent a child could delay gratification serves largely as a measure or “symptom” for how stable his home environment was, or how well his cognitive abilities were developing in the growing up phase. It was less effective as a predictor of future success.


The Need for Psychological Assessments

Rather than depend narrowly on this one skill (to delay gratification), the researchers of the replicated study emphasise a need to take into consideration the broader elements of a child’s life. 

This approach parallels to how we would study the actual work performance of an individual. Observable workplace behaviour is a function of five factors:

  1. Skill Level (Vocational and Soft)
  2. Job Role Perception
  3. Motivational factors
  4. Family & Personal Variables
  5. Psychological traits and behavioural tendencies

Therefore, it cannot be attributed solely to a particular factor. For instance, comparing two equally skilled and qualified persons –  a sole breadwinner and a person from a wealthy background. Given circumstances where the sole breadwinner is forced to succeed, it is highly likely that he will outperform the rich person. This illustrates the importance of the 4th factor – family and personal variables, that was unfortunately overlooked in the original Marshmellow research. 

The 5th factor – psychological traits and behavioural tendencies, is yet another elusive but important factor in impacting actual workplace behaviour. Feel free to catch up on our last article “Talented but not liking it? The psychology behind it”  to know more. 


Clarity is Key

As we have realised over time, the original Marshmallow test could be more for surfacing “symptoms” rather than a predictor of future success. Rather than to dive into cultivating delayed gratification as a skill per se, it would help to understand what affected the child’s ability to delay gratification in the first place. 

Similarly, when someone receives a personality assessment result stating that he is aggressive/dominant for instance, does it imply a weakness that he has to manage soonest?  Or could it be a symptom of underlying tendencies that he first needs to be aware of? 

In other words, how certain is he that the assessment mechanism is robust enough to differentiate between these tendencies in a person:  “Assertive” Vs “Aggressive/ Dominant”?

Through the SQI Traits Analysis, we assess and differentiate positive/neutral traits from negative tendencies that result from a combination of various neutral traits. These include but are not limited to the following:

          “Assertive” Vs “Aggressive/ Dominant”

          “Persevere” Vs “Stubborn”

          “Cautious” Vs “Risk-Averse”

With that, people can look forward to effective solutions based on sharper and deeper understanding of the underlying root causes and behavioural effects of people issues in a volatile, uncertain and globalised work environment.

Looking back at the original Marshmallow test, its limitations were aplenty. Yet, so were the insights and lessons learned. Just as lightbulb inventor Thomas Edison aptly put it: “I haven’t failed — I’ve just found 10,000 that won’t work.”




Revisiting the Marshmallow Test, First Published May 25, 2018

Meilan Solly, SMITHSONIANMAG.COM,Why Delayed Gratification in the Marshmallow Test Doesn’t Equal Success

Hilary BrueckBusiness Insider US, 2017, The famous Stanford ‘marshmallow test’ suggested that kids with better self-control were more successful. But it’s being challenged because of a major flaw


Leave a Reply

Your email address will not be published. Required fields are marked *

Call Now ButtonCall Us